基于特征处理与径向基神经网络的锂电池剩余容量估算方法

doi:10.19799/j.cnki.2095-4239.2020.0314

摘要/Abstract

摘要：

为解决锂电池可用容量估算过程中精度与效率难以兼顾的问题，本文提出了一种基于特征处理与径向基神经网络的锂电池剩余容量估计方法。首先由电池充电过程数据中提取与剩余可用容量相关联的特征量，然后运用局部异常因子算法对特征量中异常点进行精准清洗，提高特征量所含有效信息量，再通过局部线性嵌入降维算法对所得特征向量组进行降维处理，减少数据复杂度，最后，引入径向基神经网络建立起剩余容量的估算模型。在不同型号电池上应用该模型进行了验证，估算结果的最大平均绝对误差为0.06，最大均方根误差为0.05，表明该模型能够有效估计锂电池的剩余可用容量并有较强的鲁棒性。与Elman神经网络和BP神经网络算法相比，在保证高精度的同时该方法有更快的估算效率。

关键词: 锂离子电池, 特征处理, 径向基神经网络, 容量估计

Abstract:

To solve the problem of the difficulty in balancing the accuracy and efficiency in the process of capacity estimation for an LIB, this paper proposes a remaining capacity estimation method for an LIB based on feature engineering and a radial basis neural network. First, the features associated with the remaining available capacity from the data during battery charging is extracted; then, the local anomaly factor algorithm is used to clean the abnormal points accurately in the features that increase the amount of effective information contained in the feature quantity; next, the dimensionality reduction process of the feature vector group is performed by the local linear embedding dimensionality reduction algorithm to reduce the computation complexity; and finally, a radial basis function neural network is introduced to establish an estimation model for the remaining capacity. The model is verified on different batteries; the results show that the model has strong robustness, the maximum average absolute error does not exceed 0.06, the maximum root mean square error is 0.05, and, when compared with the Elman neural network and the BP neural network algorithm, it has faster estimation efficiency while ensuring high accuracy.

Key words: lithium-ion battery, feature processing, RBF neural network, capacity estimation

中图分类号:

TM 911

陈峥, 李磊磊, 舒星, 沈世全, 刘永刚, 申江卫. 基于特征处理与径向基神经网络的锂电池剩余容量估算方法[J]. 储能科学与技术, 2021, 10(1): 261-270.

Zheng CHEN, Leilei LI, Xing SHU, Shiquan SHEN, Yonggang LIU, Jiangwei SHEN. Efficient remaining capacity estimation method for LIB based on feature processing and the RBF neural network[J]. Energy Storage Science and Technology, 2021, 10(1): 261-270.

图/表 18

图1

表1

图2

图3

图4

图5

图6

表2

图7

图8

图9

图10

表3

图11

表4

表5

图12

图13

参考文献 23

1	TAGADE P, HARIHARAN K S, RAMACHANDRAN S, et al. Deep Gaussian process regression for lithium-ion battery health prognosis and degradation mode diagnosis[J]. Journal of Power Sources, 2020, 445: doi: 10.1016/j.jpowsour.2019.227281.
2	WANG X, WEI X, DAI H. Estimation of state of health of lithium-ion batteries based on charge transfer resistance considering different temperature and state of charge[J]. Journal of Energy Storage, 2019, 21: 618-631.
3	WASSILIADIS N, ADERMANN J, FRERICKS A, et al. Revisiting the dual extended Kalman filter for battery state-of-charge and state-of-health estimation: A use-case life cycle analysis[J]. Journal of Energy Storage, 2018: 73-87.
4	XUE Z, ZHANG Y, CHENG C, et al. Remaining useful life prediction of lithium-ion batteries with adaptive unscented kalman filter and optimized support vector regression[J]. Neurocomputing, 2019: 95-102.
5	韦海燕, 安晶晶, 陈静, 等. 基于改进粒子滤波算法实现锂离子电池RUL预测[J]. 汽车工程, 2019, 41(12): 1377-1383.
	WEI Haiyan, AN Jingjing, CHEN Jing, et al. Prediction of lithium-ion battery based on improved particle filtering algorithm[J]. Automotive Engineering, 2019, 41(12): 1377-1383.
6	郑涛, 张里, 侯杨成, 等. 基于自适应CKF的老化锂电池SOC估计[J]. 储能科学与技术, 2020, 9(4): 1193-1199.
	ZHENG Tao, ZHANG Li, HOU Yangcheng, et al. SOC estimation of aging lithium battery based on adaptive CKF[J]. Energy Storage Science and Technology, 2020, 9(4): 1193-1199.
7	GUO P, CHENG Z, YANG L. A data-driven remaining capacity estimation approach for lithium-ion batteries based on charging health feature extraction[J]. Journal of Power Sources, 2019, 412: 442-450.
8	WANG J S, LIU P, HICKSGARNER J, et al. Cycle-life model for graphite-LiFePO4 cells[J]. Journal of Power Sources, 2011, 196(8): 3942-3949.
9	ZHANG Y, XIONG R, HE H, et al. State of charge-dependent aging mechanisms in graphite/Li(NiCoAl)O2 cells: Capacity loss modeling and remaining useful life prediction[J]. Appl Energy, 2019, 255(1): doi: 10.1016/j.apenergy.2019.113818.
10	TANG X, LIU K, WANG X, et al. Model migration neural network for predicting battery aging trajectories[J]. IEEE Trans Transport Electrification, 2020: 363-374.
11	DAI H, ZHAO G, LIN M, et al. A novel estimation method for the state of health of lithium-ion battery using prior knowledge-based neural network and Markov chain[J]. IEEE Trans Ind Electron, 2019, 66(10): 7706-7716.
12	LIU K, LI Y, HU X, et al. Gaussian process regression with automatic relevance determination kernel for calendar aging prediction of lithium-ion batteries[J]. IEEE Trans Industrial Inf, 2020, 16(6): 3767-3777.
13	LIU K, SHANG Y, OUYANG Q, et al. A data-driven approach with uncertainty quantification for predicting future capacities and remaining useful life of lithium-ion battery[J]. IEEE Trans Ind Electron, 2020, doi: 10.1109/TIE.2020.2973876.
14	起文斌, 张华, 金周, 等. 锂电池百篇论文点评(2020.04.01—2020.05.31)[J]. 储能科学与技术, 2020, 9(4): 1015-1029.
	QI Wenbin, ZHANG Hua, JIN Zhou, et al. Reviews of selected 100 recent papers for lithium batteries(Apr. 01, 2020 to May 31, 2020)[J]. Energy Storage Science and Technology, 2020, 9(4): 1015-1029.
15	LI Y, ABDEL-MONEM M, GOPALAKRISHNAN R, et al. A quick on-line state of health estimation method for Li-ion battery with incremental capacity curves processed by Gaussian filter[J]. Journal of Power Sources, 2018, 373: 40-53.
16	WANG D, KONG J Z, ZHAO Y, et al. Piecewise model based intelligent prognostics for state of health prediction of rechargeable batteries with capacity regeneration phenomena[J]. Measurement, 2019, doi: 10.1016/j.measurement.2019.07.064.
17	BREUNIG M M, KRIEGEL H P, NG R, et al. LOF: Identifying density-based local outliers[J]. ACM Sigmod Record, 2000, 29(2): 93-104.
18	RICHARDSON R R, OSBORNE M A, HOWEY D A. Battery health prediction under generalized conditions using a Gaussian process transition model[J]. Journal of Energy Storage, 2019, 23: 320-328.
19	ZHOU Y, HUANG M, CHEN Y, et al. A novel health indicator for on-line lithiumion batteries remaining useful life prediction[J]. Journal of Power Sources, 2016, 321: doi: 10.1016/j.jpowsour.2016.04.119.
20	ROWEIS, SAM T, SAU L, et al. Nonlinear dimensionality reduction by locally linear embedding[J]. Science, 2000: 232-2326.
21	LU C, TAO L, FAN H. Li-ion battery capacity estimation: A geometrical approach[J]. Journal of Power Sources, 2014, 261(9): 141-147.
22	OLGA K, OLEG O, MATTI P. Selection of the optimal parameter value for the locally linear embedding algorithm[C]//Scandinavian Conference on Image Analysis, 2002.
23	ER M J, WU S, LU J, et al. Face recognition with radial basis function (RBF) neural networks[J]. IEEE Transactions on Neural Networks, 2002, 13(3): 697-710.

电池编号	特征量
电池编号	CC-T	CV-T	CC-C	CV-C	CC-T/CV-T	CC-C/CV-C	max-IC
电池A-1	0.991/0.997	-0.985/-0.992	0.992/0.992	-0.955/-0.962	0.987/0.995	0.972/0.974	0.996/0.997
电池B-2	0.975/0.998	-0.967/-0.976	0.996/0.996	0.971/0.998	0.962/0.993	0.983/0.991	0.991/0.993
电池B-3	0.980/0.999	-0.954/-0.988	0.980/0.999	0.937/0.999	0.956/0.997	0.955/0.973	0.989/0.994
电池B-4	0.986/0.998	-0.936/-0.985	0.985/0.988	0.951/0.996	0.945/0.994	0.819/0.832	0.998/0.999

电池标号及训练量	电池A-1			电池B-2			电池B-3			电池B-4
电池标号及训练量	50%	60%	70%	50%	60%	70%	50%	60%	70%	50%	60%	70%
MAE	0.0475	0.0370	0.0327	0.1067	0.0460	0.0299	0.0292	0.0258	0.0239	0.0646	0.0494	0.0381 0.0022
MSE	0.0026	0.0018	0.0014	0.0137	0.0031	0.0011	0.0018	0.0012	0.0008	0.0056	0.0034	0.0381 0.0022
RMSE	0.0506	0.0428	0.0376	0.1171	0.0557	0.0330	0.0429	0.0343	0.0290	0.0748	0.0581	0.0472

方法	MAE	MSE	RMSE	Time/s
RBF NN	0.0612	0.0043	0.0570	0.99
Elman NN	0.0847	0.0076	0.0874	3.51
BP NN	0.0730	0.0070	0.0834	8.54

方法	MAE	MSE	RMSE	Time/s
RBF NN	0.0439	0.0021	0.0463	0.90
Elman NN	0.0518	0.0051	0.0714	3.59
BP NN	0.0643	0.0076	0.0885	9.45

[1]	李海涛, 孔令丽, 张欣, 余传军, 王纪威, 徐琳. N/P设计对高镍NCM/Gr电芯性能的影响[J]. 储能科学与技术, 2022, 11(7): 2040-2045.
[2]	刘显茜, 孙安梁, 田川. 基于仿生翅脉流道冷板的锂离子电池组液冷散热[J]. 储能科学与技术, 2022, 11(7): 2266-2273.
[3]	陈龙, 夏权, 任羿, 曹高萍, 邱景义, 张浩. 多物理场耦合下锂离子电池组可靠性研究现状与展望[J]. 储能科学与技术, 2022, 11(7): 2316-2323.
[4]	易顺民, 谢林柏, 彭力. 基于VF-DW-DFN的锂离子电池剩余寿命预测[J]. 储能科学与技术, 2022, 11(7): 2305-2315.
[5]	祝庆伟, 俞小莉, 吴启超, 徐一丹, 陈芬放, 黄瑞. 高能量密度锂离子电池老化半经验模型[J]. 储能科学与技术, 2022, 11(7): 2324-2331.
[6]	孔为, 金劲涛, 陆西坡, 孙洋. 对称蛇形流道锂离子电池冷却性能[J]. 储能科学与技术, 2022, 11(7): 2258-2265.
[7]	霍思达, 薛文东, 李新丽, 李勇. 基于CiteSpace知识图谱的锂电池复合电解质可视化分析[J]. 储能科学与技术, 2022, 11(7): 2103-2113.
[8]	邓健想, 赵金良, 黄成德. 高能量锂离子电池硅基负极黏结剂研究进展[J]. 储能科学与技术, 2022, 11(7): 2092-2102.
[9]	王宇作, 王瑨, 卢颖莉, 阮殿波. 孔结构对软碳负极储锂性能的影响[J]. 储能科学与技术, 2022, 11(7): 2023-2029.
[10]	丁奕, 杨艳, 陈锴, 曾涛, 黄云辉. 锂离子电池智能消防及其研究方法[J]. 储能科学与技术, 2022, 11(6): 1822-1833.
[11]	张言, 王海, 刘朝孟, 张德柳, 王佳东, 李建中, 高宣雯, 骆文彬. 锂离子电池富镍三元正极材料NCM的研究进展[J]. 储能科学与技术, 2022, 11(6): 1693-1705.
[12]	欧宇, 侯文会, 刘凯. 锂离子电池中的智能安全电解液研究进展[J]. 储能科学与技术, 2022, 11(6): 1772-1787.
[13]	韩俊伟, 肖菁, 陶莹, 孔德斌, 吕伟, 杨全红. 致密储能：基于石墨烯的方法学和应用实例[J]. 储能科学与技术, 2022, 11(6): 1865-1873.
[14]	辛耀达, 李娜, 杨乐, 宋维力, 孙磊, 陈浩森, 方岱宁. 锂离子电池植入传感技术[J]. 储能科学与技术, 2022, 11(6): 1834-1846.
[15]	燕乔一, 吴锋, 陈人杰, 李丽. 锂离子电池负极石墨回收处理及资源循环[J]. 储能科学与技术, 2022, 11(6): 1760-1771.